Basic Photoshop
Basic Photoshop
realize there was a handy feature to automatically stitch your panoramic images together. What you may not realize is you can also use this feature to create a photo montage. If you have a series of panoramic images, Photomerge works great. From the File>Automate menu, select Photomerge.
This will launch the browser so you can find and select the images you want to stitch together.
When you have selected all of the image, click OK and let Photoshop do its thing. It should only take a moment (depending on the file size and number of images). Photoshop looks through all of the images, and searches for common edges and shapes and attempts to stitch them together.
It does a pretty good job with most images, but you may find you need to tweak it a bit. It is not as good as having an application like Stitcher from RealViz, but it is pretty close. One of the options in Photomerge to pay special attention to is the Advanced Blending Feature. This feature attempts to blur and blend the overlapping images together to create the seamless final composite. Unfortunately there are not advanced controls to determine how much blending to do, but the built in algorithms seems to work nearly every time. But what if you dont want to create a panoramic image? What if instead you wish to quickly create a montage of photos for a magazine spread, website, or poster? You can use Photomerge for that too. Simply import the images you wish to use in your montage. Photoshop CS will not be able to detect any matching images, so it will default to a blank canvas with all of the images at the top of the screen.
Now drag your images from the bar to the canvas and overlap as needed. If you need to rotate images, use the Rotate Tool in the Photo merge interface.
When you save the file you have the option of saving it as is using the Photo merge blending option, or saving it with layers. The layers method places each image on a separate layer and allows you to create your own blending method.
While this may not be the best method for creating a final montage of images, it is certainly a down and dirty method for quickly generating a concept to show a client for fast approval
Automating Image Mergers in Photoshop CS2 Creating the action In order to create our action, we're simply going to do everything we did in the previous installment in this tutorial series, which you can read by clicking here. The difference is that while we're doing it, we're going to have our Actions palette open so that each step can be recorded for later using in a batch processing operation. If you don't want to know how to create your own actions for merging documents, be sure to skip forward to the last section in this article ("Bonus Download") to download a set of seven complete preset actions. The exact way you'll create you action depends on the sort of document merging you want to perform. The method we discuss here and the options for each step assume you want to maintain the layer structure of all of the documents you merge. And it assumes you want to merge the documents into a single new image file. Begin by creating a new document at the final desired dimensions for your project. Make sure it's called "Untitled 1" or some other name that you will use any time you run this batch process. For this procedure, the name must be consistent, or the batch won't work properly. (The names of the individual documents that will be merged don't matter.) Then open up a sample document that contains multiple layers. It doesn't really matter which document. We're just using it for the creation of our action. Now it's time to create the action itself. Open up you Actions palette. Create a new set by clicking the folder icon at the bottom of the Actions palette. I'll name my set "Nagel Series 3 Actions (Merging)," since that's the name of the set of actions I'm providing for download later on.
Then add a new action tot he set by clicking the new action button at the bottom of the Actions
palette.
The action is now recording, and everything you do from this point on will be recorded in the action. So if you need to do something outside the scope of this action, be sure to pause the recording by hitting the Stop button at the bottom of the Actions palette. Now all we need to do is work quickly through each step we want included in the action, which will involve creating a group out of our existing layers, and duplicating that group to the document entitled "Untitled 1." (Remember, each of these steps is covered in detail in our previous tutorial, which you can read by clicking here in case you need additional information.) 1. Begin by typing Option-Command-A (Mac) or Alt-Control-A (Windows) to select all of the layers in the current document.
2. Now type Command-G (Mac) or Control-G (Windows) to create a new group from the layers you've selected. (You can skip this step, if you don't want to use layer groups. You can also, instead, choose to merge the layers into a Smart Object by choosing Layer > Smart Objects > Group into New Smart
Object.)
3. Then choose Layer > Duplicate Group. If you did not group your layers in Step 2, then choose Layer > Duplicate Layers instead. If you created a Smart Object from your layers in Step 2, choose Layer > Duplicate Layer now. In the dialog that pops up, set the target to "Untitled 1."
4. Finally, close your image without saving. This adds the final step to your action, and you can now stop recording.
When you stop recording, you can then close Untitled 1 as well. Batch processing with your new action Now, the action we just created will allow us to open up multiple documents and merge them into a single new document automatically. Begin, once again, by creating a new document called "Untitled 1" that will be used as the target for the document merger. Set the document's dimensions and resolution to whatever you with to use for your final project. You're no ready to run your batch operation. 1. Choose File > Automate > Batch. 2. In the dialog that pops up, set the action to the one you just created. And also choose the folder where your source images reside. Do not set a destination folder, as you're not saving any images; you're just creating one new merged image, which you can save when the batch process is complete.
3. Let the batch run its course. When it's done, you'll wind up with a document that contains all of the original documents, grouped neatly into their own little layer groups. In the example below, you see that I batch processed three separate images to merge them into a single image.
You will also note that all the groups have the same name. You can rename these by double-clicking the names in the Layers palette. Now you can save your merged document, and you're done. Shadow/Highlight inside Photoshop CS part 2 With the Highlight value, the same rules apply, so there's no need to go into those. The next parameter down, Color Correction allows you to make the colors in the corrected regions more or less vivid. (On grayscale images, a Brightness adjustment is available instead.) The examples below show varying degrees of Color Correction. The default is +20, which you've seen in all of the previous images. (All of the examples below use Shadow and Tonal Width values of 100 and a Radius of 30.)
Now obviously when you're pulling out detail from the dark areas of an image, some of that detail is bound to be in the form of noise or artifacts. For example, in the image below, when I crank up the Shadow value to 100 percent, I see strong horizontal noise emerging from the image.
So how can this be alleviated? How can you get the detail without so much noise? Here's how. 1. Convert your background to a layer by double-clicking it in the Layers palette. Then duplicate that layer.
3. In the Layers palette, move the original bottom layer up to the top of the stacking order.
4. With the top (darker) layer selected, switch its blending mode from Normal to either Soft Light or Pin Light. And voila!
But what if I don't want it to be in silhouette? What if I want to bring out the details on the mountain and the foreground from this excessively dark photograph? No problem.
How'd I do that? Well, I just selected Image > Adjustments > Shadow/Highlight, and cranked up the "Shadow" value to 100 percent.
And that's literally all there was to it. It sucked out the data that existed in the file (despite resizing and multiple JPEG compressions) and composited it into the original in real time as I adjusted the slider, similar to much more painstaking processes that probably wouldn't yield results nearly as satisfactory. But Shadow/Highlight is not just about shadows. It's an image adjustment function designed to accentuate the details in highlights as well. So here's the same image with a 77 percent Highlight adjustment to provide a little definition in the clouds. (Note the differences in the cloud formation on the right.)
Not too bad? Well, of course, I don't want to give you the idea that this is some kind of magical image manipulation tool. If there's detail in your photograph, Shadow/Highlight can find and accentuate it. But in images that are too overexposed, detail can be permanently lost. To a lesser degree, the same goes for images that are too underexposed. In cases where your lighting simply doesn't allow for exposures that capture data in the extremes of light and dark, you'll still be better off bracketing your photos and then using something like the Blend Exposure filter from Reindeer Graphics' Optipix suite. But even with difficult images, Shadow/Highlight provides additional tools for finessing all the data out of your image. Beyond its simple slider adjustments for Shadow and Highlight values, the Shadow/Highlight function also provides a wide variety of tools for fine-tuning your adjustments. You can access these by clicking on the "Show More Options" checkbox at the bottom of the dialog.
The Tonal Width settings allow you to to tighten or widen the range of tones that will be affected by your Shadow/Highlight adjustment. For example, you may want to bring out the details in t=only the darkest portions of your image, leaving some of the midrange elements alone, or you may want to
brighten up more than just the darkest portions. The images below show examples of variations on the Tonal Width setting and their impact on neighboring tonal values.
Radius, on the other hand, does not target tonal values but instead allows you to adjust the area around each pixel in the adjustment. It's a bit more obscure than Tonal Width, but it can have a dramatic impact on the final result of your manipulation. The images below show variations on Radius. (Each uses a Shadow and Tonal Width value of 100.)
For this example, we're going to create a simple window frame--just six rectangles--and we're going to start by using the rectangular Marquee tool. While you can also easily do this using the rectangular Path tool, I'd like to show you how to convert a selection into a path so that you might get a little more out of this. 1. Begin by using your rectangular Marquee tool to select an area that will become the basic shape of one individual window pane.
2. Open up your Path palette, and choose the flyaway menu located on the top right of the palette. From the menu, select "Make Work Path." This will convert your selection into a rectangular path.
3. In your Tool palette, switch over to the Path Selection tool, which is the tool that looks just like a regular cursor. When you've chosen it, use the tool to click on the rectangular path in your canvas window.
4. From the Edit menu, choose Copy, or type Command-C. Then Choose Paste from the Edit menu, or type Command-V. You want to paste five times, giving you a total of six individual rectangular paths. However, when you paste them, they will appear on top of one another, looking as if you still had only one path. Not to worry. Simply use your Path Selection tool to move the paths one by one into position, as seen below. You can hold down the shift key while you're moving the paths in order to constrain the paths along the horizontal or vertical axes.
5. Now use your Path Selection tool to select all of the little rectangles at once. You can do this by hold down the Shift key and clicking on them individually, or you can draw a Marquee around all of the paths to select them all at once.
6. Now, when you have all of the paths selected at once, you'll see an option up in your top tool bar called "Combine." Click this button to combine your six individual paths into a single path containing six rectangles.
7. Finally, from the Edit menu, choose "Define Custom Shape...." Give it a name that describes the shape for easy identification later. You might want to create a number of variations on this for future use.
Now, what is the benefit of doing this? Well, not only will you never have to draw this particular shape manually again, but you will also be able to apply this exact shape in a variety of sizes (interactively) and using automatic fills and strokes. You will also be able to transform the shape (scale, perspective, etc.), without losing resolution. Your custom object is now accessible via the Custom Shape tool. When you select the Custom Shape tool, you will then be able to select any shape you've created from the shape pull-down menu located in your top Tool Bar.
Here's how it looks in action with a black foreground color and "Shape Layers" option selected in the top Tool Bar.
Set to Color Mode and placed over another image, you get this:
Color Mode is pretty close to Hue Mode, so lets look at Hue Mode again to get rid of any confusion. Hue Mode keeps the brightness and satuation of the base color but applies the hue of the blend layer. Color Mode keeps the brightness of the base color and applies both the hue and saturation of the blend color. So heres how the scenario changes when the blue/red layer is set to Hue Mode instead:
As you can see, theres more grey in this picture; the blue and reds are desaturated in the areas where the original image didnt have much saturation. Does that make more sense? And following Photoshops practical application advice, here is a sepia-toned cherry blossom:
Saturation Blend Mode in Photoshop Were in the home stretch of the Photoshop Blending Modes series (good heavens, that means I have to come up with more topics to write about!), and todays blog looks at the Saturation Blend Mode. Recall how the Hue blending mode kept everything the same on the base layers, but applied the hue (color) from the blend layer. Saturation mode is similar, where everything stays the same on the base layer, except the saturation from the blend layer is applied. Try this example on for size: Start with my cherry blossom picture
Add this blend layer, which has a highly-saturated blue and a not-very-saturated blue:
As expected, the top half (the light, not-very-saturated blue) results in a less-saturated flower image, while the bottom half is very saturated. Now for a practical application. Ill start with my standard rose image and then add a radial gradient to produce this blend layer, using a not-very-saturated color:
(To do this, I had to open the Gradient Editor by clicking on the gradient preview in the top toolbar, and make a new gradient from the foreground-to-transparent gradient, then flip the sliders that defined the gradients to make a transparent-to-foreground gradient.)
Combining the radial gradient blend layer (set to Saturation mode) results in this image:
It kind of reminds me of a breathing life into something kind of image, maybe representing some kind of inner life or inner glow. Im sure you can find your own use for this technique
Ill put a solid blue layer over it and set it to Hue mode. The rose layer will be colorizedkind of with the blue (but note that the areas around the rose have a bit more grey than blue in them because
So, heres yet another way to turn a color image into black and white: Add a blackor whitelayer and set it to Hue mode:
Setting a new, blank layer to Hue means that you can paint over any part of your image and change the color. As a kind of extreme example, I made a selection of just the rose and then filled it with a rainbow gradient.
Photoshops describes Vivid Light blending mode as combining the effects of Color Dodge and Color Burn modes like this: Vivid Light: Burns or dodges the colors by increasing or decreasing the contrast, depending on the blend color. If the blend color (light source) is lighter than 50% gray, the image is lightened by decreasing the contrast. If the blend color is darker than 50% gray, the image is darkened by increasing the contrast. In reality, however, Vivid Light brightens images similar to Color Dodge mode, but doesnt darken quite as much as Color Burn. In this diagram below, I created two layers, one with a light grey box, one with a dark grey box, and set the former to Color Dodge and the latter to Color Burn. You can see that the Color Dodge layer lightens the image dramatically and the Color Burn layer darkens the image dramatically. But in the second sample, I took the two boxes, put them on the same layer, and set it to Vivid Light. The light box does result in a lighter image comparable to Color Dodge, but the dark box has significantly less contrast as compared to Color Burn. The thing to remember from all of this is that Vivid Light mode will take the blend layer and make the lighter areas a lot brighter than it will make the dark areas darker.
For my practical application, Ill use Vivid Light mode to overlay text on top of other images. Because Vivid Light doesnt really show the blend layer, but just uses the blend layer to affect the colors of the base layer, the text will look more like its part of the picture, as opposed to being a slightly transparent layer on top of a picture.
Then Ill add my semi-standard pawprint layer and set it to Color Dodge:
The pure black area of the pawprint layer results in no change, but all other areas dramatically lighten the image underneath.
So I have to admit that Color Dodge mode is one that Ive never used in an actual application before. But apparently a few other people in Google-world have used Color Dodge mode for lots of different purposes. A lot of people will use Color Dodge layers when making high-tech or grunge-style abstract images. But there were two other suggestions that I found interesting
In this example, I started with a blue-to-white gradient layer, then overlaid it with a black-to-white gradient layer. The black-to-white layer was set to the color burn blending mode. (The result shows a strip of the original black-to-white layer for your comparison convenience.) This example demonstrates the increasing the contrast aspect of color burning. The darkness of the black-to-white layer doesnt do much to the part of the gradient that is pure blue, but it does affect the faded part of the blue layer by giving it more contrastmaking it more blue. Example 2
In this example, I took the same blue-to-white base layer and overlaid it with a fuschia-to-white layer set to color burn. While Im not completely sure what Photoshop means when it talks about reflecting the blend color, perhaps this is it: again, the left pure blue area is pretty much untouched, but the fuschia both increases the saturation of the lighter blue areas and also blends in a little bit. Example 3
This example takes a rainbow layer and overlays it with a black-to-white and rainbow layer set to color burn. The result may be surprising! Looking at where the rainbow overlaps with the greyer parts of the black-and-white gradient makes sense to methe contrast of the rainbow is definitely heightened by the grey gradient. Its the other parts that may be unexpected. But what I like about this diagram is that it does show me what to expect when I burn colors onto other colorsfor example, red is not really affected by any other color (is it a coincidence that red is associated with hot?). On to my easy practical application. Lets start with this picture of a dog:
Its a cute dog, but the picture seems a little bit dull. So, I add a new layer on top, set it to Color Burn, and start painting on it with grey around the dog. (I set the opacity to be a little bit less so that the darkening effect wasnt so stark.) I use different shades of grey to get the various patches of grass at about the same level of darkness.
And after only two minutes of manipulation, the dog really stands out!
I like using Linear Light to overlay text (or line art) on top of an image for a neat watermark effect. I first add text in a medium grey color (#939393):
I set this top layer to Linear Light and get this result:
Now, I dont pretend to fully understand exactly how the Multiply Mode works when it comes to blending colorsI kind of get it, but I wouldnt be able to predict what the end result of blending two images or colors would look like to great detail. Youll probably find as well that its just going to take some experimentation. Maybe this example will help: Take the original image of the dog, plus a rainbow pawprint. Set the pawprint to Multiply mode, and heres the resultkind of like a clear rainbow plastic effect, but that the result colors are slightly darker, as well:
Okay, now for the practical applications. I have two: 1. Darken pictures. This application is sometimes useful. In particular, when you have a slightly overexposed photo (think washed out), you can duplicate the photo on a new layer and set it to multiply, then play with the opacity level of the new layer until you feel the darker values are adequately dark.
Ill duplicate the layer, set it to Multiply mode, and change the opacity to 40 or 50%:
Heres the result; while the light areas havent been too affected, the darker areas are definitely a little darker:
When I layer the two images on top of each other and set the top layer to the Darken mode, youll notice that the darker values show from each layer.
But the effect is even cooler when you start creating a montage with different photos or illustrations that have a lot of colors. Lets take these two flower photos:
Once again, layering the images and setting the top layer to the Darken mode creates an interesting effect:
Because the cherry blossom picture has an area that is very light (the blossom), surrounded by a fairly dark background (the tree trunk), the yellow rose part shows through in a pretty defined area. I like it! Just to demonstrate that Photoshop is really taking the darker value for each pixel, Ill switch up the layers so that the rose layer is on top, set to Darken Mode. The end result looks exactly the same!
(And a note for those who care: Im using Photoshop 7.) Just a quick intro to blending modes for those of you who havent tried using them before: 1. Blending modes need two or more layers for any effect to workyou need at least two layers to blend together. 2. A layer with a blending mode specified will blend with all other colors from layers underneath it. We call those original layer colors the base colors. The colors on the top layer (with the blending mode selected) is called the blend colors. The resultwhat you see as a result of the blendingis called the result color. And finally: Dissolve Mode According to Photoshop Help: Dissolve Mode edits or paints each pixel to make it the result color. However, the result color is a random replacement of the pixels with the base color or the blend color, depending on the opacity at any pixel location. Say what? Worded in a way that makes more sense to me, and maybe you, too: The result of Dissolve Mode looks speckled, kind of like when a PowerPoint slide, or movie scene, dissolves into another (hence the name!). Dissolve Mode only produces this result when there are partially transparent pixels by taking the partially transparent pixels and randomly choosing some to display at 100% opacity for a speckled effect. (The higher the original opacity, the more solid the result is.) This is where an example comes in handy. Lets start with a picture of a bouquet of flowers that I got for Valentines Day:
Now, Ill add a custom shape layer and set the blending mode to Dissolve:
Hmmm, doesnt look much different from the Normal mode! Thats because this layer doesnt have any partially transparent pixels. When I set the opacity to 60%, the Dissolve blend does its magic:
Note that Photoshop takes those 60% pixels and displays them, randomly, either at 0% or 100%. If I change the opacity to 20%, the resulting image shows less speckles from the pawprint layer:
Duplicating the layer and setting the top layer to Linear Burn results in an overall darker picture, although the darker areas are more dark.
Now for the practical application. Ill start with these two (cropped) images:
Putting the leaf layer on top, I set it to Linear Burn, and this is what I get:
The Linear Burn example is slightly darkerlinear burn tends to give you more blacks than multiply mode would. Either effect is useful for me, it just comes down to what mood Im in that day!
Choose the Pen Tool. In the horizontal options bar, make sure that "Paths," not "Shape layer," is selected. Then, start using the Pen Tool to create a path around the object. I find that it's helpful to zoom in for greater detail.
Here's my completely outlined bee. You can see all the different points that I created using the Pen Tool, as well as the resultant path.
2.
contrast with the object (set a high tolerance and use one click!), but may involve a bit of touch-up work at the end. No good for objects that are on multi-colored backgrounds. The Background Eraser Technique This technique would be great on a photo object that is already "isolated" but has a solid, contrasting background color. Even still, I might pass up this technique for the next method, as they're essentially the same and I think the Extract method goes a little faster. The Extract Filter Technique Works best for erasing mostly solid-colored, contrasting backgrounds. Photoshop also does a nice job of getting rid of the color halo problem but you may have to do some additional touch-up work at the end. The Quick Mask Technique Almost the ideal solution: relatively quick, pixel-perfect control over edges, and works with any kind of background. Plus, by using a layer mask, you don't actually delete any part of the background. The only complaint I have with this method is that you may get a color halo. The Pen Tool Technique Fantastic technique for objects with hard edges, and it works with any kind of background. You can be as detailed as you want to get a perfect outline, and you can always go back and modify the path if you need to. Plus, you get lots of practice using the Pen Tool.
3.
4.
5.
If you've never used Quick Mask mode in Photoshop, you're missing out on a huge time-saver! Quick Mask mode lets you use brush, pencil, paint bucket, etc. tools to paint an "inverse selection" using shades of grey and black. (Black identifies parts that won't be selected; grey identifies parts that, when selected, will have some level of transparency.) When you switch back to Normal mode, the "clear" parts of your Quick Mask are selected so that you can apply changes. We'll use Quick Mask mode to paint the background of the image, then apply a Layer Mask to hide the background. Click the Quick Mask icon in the tool palette to switch to Quick Mask mode.
Using a fairly large brush, with the foreground color set to black, I'll start to block out the background. You'll notice that the masked areas in which you paint will turn pink.
The large brush size allows me to color in the main areas of the background fairly quickly.
I can then use smaller brushes to fill in the details around the edges.
Tip: You may change the foreground color to white if you want to "paint back" parts of the object. Switch back to Normal Mode, and you'll see the selection. Because everything that's painted in is not selected, the final result is that the bee is selected.
At this point, I click the Layer Mask icon at the bottom of the Layers Palette to create a layer mask, which essentially isolates the bee from the background.
And in summary: Quick Mask Technique: 1. 2. Switch to Quick Mask mode and paint everything you want to get rid of. Switch back to normal mode and use the resultant selection to make a Layer Mask.
My take: A great method, because it gives you precision-control over what gets selected. This method works well for any picture -- solid background color or not. The only unfortunate aspect of this method is that, for some objects, you may see a "color halo" where the edges have picked up on some of the background color. (One way to get around this is to take a little more time and paint around the edges with a grey airbrush to make the edges a little transparent.) Note: This same type of effect can be achieved by creating a Layer Mask, then painting on the layer mask with black paint to hide the background. The two techniques are virtually identical -- I just chose to write about Quick Mask because I've already covered the concept of Layer Masks in another article!
As with the Background Eraser technique, it's a smart idea to set the source for the History Brush early in the procedure, in case you need to do some touch-up later on.
Now, let's run the Filter > Extract command. (In earlier versions of Photoshop, this was located under the Image menu.)